R: Error Measures for Estimated Marginal Likelihood

error_measures {bridgesampling}

R Documentation

Error Measures for Estimated Marginal Likelihood

Description

Computes error measures for estimated marginal likelihood.

Usage

error_measures(bridge_object, ...)

## S3 method for class 'bridge'
error_measures(bridge_object, ...)

## S3 method for class 'bridge_list'
error_measures(bridge_object, na.rm = TRUE, ...)

Arguments

`bridge_object`	an object of class `"bridge"` or `"bridge_list"` as returned from `bridge_sampler`.
`...`	additional arguments (currently ignored).
`na.rm`	a logical indicating whether missing values in logml estimates should be removed. Ignored for the `bridge` method.

Details

Computes error measures for marginal likelihood bridge sampling estimates. The approximate errors for a bridge_object of class "bridge" that has been obtained with method = "normal" and repetitions = 1 are based on Fruehwirth-Schnatter (2004). Not applicable in case the object of class "bridge" has been obtained with method = "warp3" and repetitions = 1. To assess the uncertainty of the estimate in this case, it is recommended to run the "warp3" procedure multiple times.

Value

If bridge_object is of class "bridge" and has been obtained with method = "normal" and repetitions = 1, returns a list with components:

re2: approximate relative mean-squared error for marginal likelihood estimate.
cv: approximate coefficient of variation for marginal likelihood estimate (assumes that bridge estimate is unbiased).
percentage: approximate percentage error of marginal likelihood estimate.

If bridge_object is of class "bridge_list", returns a list with components:

min: minimum of the log marginal likelihood estimates.
max: maximum of the log marginal likelihood estimates.
IQR: interquartile range of the log marginal likelihood estimates.

Note

For examples, see bridge_sampler and the accompanying vignettes:
vignette("bridgesampling_example_jags")
vignette("bridgesampling_example_stan")

Author(s)

Quentin F. Gronau

References

Fruehwirth-Schnatter, S. (2004). Estimating marginal likelihoods for mixture and Markov switching models using bridge sampling techniques. The Econometrics Journal, 7, 143-167. doi: 10.1111/j.1368-423X.2004.00125.x